I recently encountered a puzzling issue with our inventory system that tracks PCs using SMB/SNMP scanning. After moving the system to a new server, it couldn't connect to some devices. Initially, we thought the Windows 11 update caused the problem, but it turned out to be a setup mistake—a team member set the wrong subnet mask (it was supposed to be /23, but they set it as /24). Once we fixed that, everything was back to normal.
However, while troubleshooting, I noticed something odd. Our network setup has an IP range of 192.168.100.0/23 with the default gateway at 192.168.101.254. The server had the IP 192.168.100.50. Surprisingly, I was able to ping the gateway, which shouldn't be possible with a /24 subnet mask. When I changed the gateway to 192.168.100.254 instead, the ping failed. This situation led me to wonder what is happening in Windows (specifically on Windows Server 2022, but I've also seen this behavior on my Windows 11 client). How does Windows handle a misconfigured default gateway? According to standard subnetting rules, this shouldn't happen. Can anyone shed some light on this?
1 Answer
It seems a bit puzzling, but here's the deal: Communication at Layer 2 is what actually allows your devices on the same switch to 'talk' to one another, even with a mismatched subnet mask. Your server can get the MAC address of the gateway through an ARP request and send data as a Layer 2 frame. That's why you could ping the gateway at 192.168.101.254 despite the apparent subnetting issue.
That makes sense! But why did changing the gateway to 192.168.100.254 cause the ping to fail?